Deep Gaze I: Boosting Saliency Prediction with Feature Maps Trained on ImageNet

TL;DR

This paper introduces a novel saliency prediction model that leverages pretrained deep neural networks for object recognition, significantly outperforming existing models and providing new insights into fixation behavior.

Contribution

It reuses pretrained deep neural networks for object recognition to improve saliency prediction, achieving state-of-the-art results and offering insights into fixation mechanisms.

Findings

Outperforms all state-of-the-art models on MIT Saliency Benchmark

Reuses pretrained networks to enhance saliency prediction

Provides new insights into fixation psychophysics

Abstract

Recent results suggest that state-of-the-art saliency models perform far from optimal in predicting fixations. This lack in performance has been attributed to an inability to model the influence of high-level image features such as objects. Recent seminal advances in applying deep neural networks to tasks like object recognition suggests that they are able to capture this kind of structure. However, the enormous amount of training data necessary to train these networks makes them difficult to apply directly to saliency prediction. We present a novel way of reusing existing neural networks that have been pretrained on the task of object recognition in models of fixation prediction. Using the well-known network of Krizhevsky et al. (2012), we come up with a new saliency model that significantly outperforms all state-of-the-art models on the MIT Saliency Benchmark. We show that the structure of this network allows new insights in the psychophysics of fixation selection and potentially their neural implementation. To train our network, we build on recent work on the modeling of saliency as point processes.